Snap Inc (SNAP.US) closed down nearly 6% on Wednesday, hitting its lowest close in more than a month and marking its second consecutive trading day of declines.
The stock has fallen nearly 5% so far this week and is on track for a second straight weekly loss.
Retail sentiment on the market sentiment tracker Stocktwits has also turned bearish.
The trigger for this sell-off did not come from earnings or regulatory developments, but rather from CEO Evan Spiegel's social media promotion of the SPECS AR glasses.
Spiegel shared an early hands-on experience with SPECS on Tuesday and emphasized that the company is expanding its university developer community.
However, the comment section quickly veered away from the product itself — some users directly asked Spiegel when he would increase his stake in the company, others asked "Would you buy it now," and one user even predicted that SPECS would become "one of the biggest failures in tech history."
Investor frustration is not hard to understand.
SPECS is priced at $2,195, with pre-orders requiring a $200 refundable deposit, and is expected to ship first in the United States, United Kingdom, and France this autumn.
The glasses offer a 51-degree field of view and up to four hours of battery life, with a companion charging case extending total usage time to 20 hours.
Snap subsequently launched a $2,395 Connected Case bundle that adds cellular connectivity, and announced commercial application partnerships with enterprise partners including NVIDIA (NVDA.US), Salesforce (CRM.US), and Amazon (AMZN.US) AWS.
In-person demonstrations will begin on October 1 in Los Angeles.
Additionally, the company is developing SPECS Intelligence — an AI assistant spanning the glasses, iPhone, and Mac platforms — with a limited preview currently available on iOS, while the Mac version remains invitation-only.
The problem is that Snap's core user base — young people and teenagers — is highly unlikely to spend over two thousand dollars on a pair of glasses.
This structural contradiction has been repeatedly raised by Wall Street since SPECS was unveiled in June.
Roth Capital Partners analyst Rohit Kulkarni maintained a "Neutral" rating and $7 price target on Snap at the time, stating bluntly that "distribution will skew toward developers, prosumers, and wealthy early adopters, with virtually no synergy with Snapchat's young and price-sensitive user base."
Management internally views 100,000 units in sales as a "stretch goal," which translates to roughly $220 million in revenue — contributing only about 1% to the consensus revenue estimate for 2026.
BNP Paribas analyst Nick Jones expressed similar concerns, believing that pricing and ongoing investment requirements will limit the product's near-term impact.
He noted that while SPECS is priced below Apple (AAPL.US) Vision Pro's starting price of $3,499, it is far above Meta (META.US) product line starting at around $250.
Just hours after Spiegel promoted SPECS, Meta countered with a full-price-band product matrix at its Connect conference, releasing an entire lineup of smart glasses spanning a wide pricing range, directly undercutting SPECS's pricing strategy.
Specifically, Meta Adventurer glasses start at $249 and go on sale October 23; camera-less Ray-Ban Meta Audio glasses start at $349 and ship October 13; and camera-equipped Ray-Ban Meta Gen 3 starts at $449 and is already available for purchase.
Meta stated that by year-end its glasses lineup will cover more than 100 styles.
Additionally, Meta previewed a VR headset priced at $1,299, targeting spring 2027 — this product uses an external computing unit and follows a completely different technical approach from SPECS's standalone all-in-one design, making direct comparisons inappropriate.
The $349 Audio version has no visual display function at all, representing a fundamental difference from SPECS.
Setting aside product form differences, Meta's market dominance in the smart glasses segment is very real.
Counterpoint Research data shows that global AI glasses shipments grew 263% year-over-year in the first half of 2026, with display-less AI glasses accounting for 96% of that total, and Meta capturing 94% shipment share in this sub-segment, with shipments up 260% year-over-year and 22% quarter-over-quarter.
This scale means Meta's advantages in supply chain bargaining power, channel distribution, and brand recognition will be difficult to challenge in the near term.
Snap's position is therefore even more delicate: it has chosen a technically more aggressive and more expensive path, attempting to stake an early claim in the "true AR" space that Meta has yet to conquer.
But the market's current feedback is that investors see neither strong support for near-term sales nor a clear path to profitability.
Snap announced layoffs of approximately 1,000 employees (16% of its global workforce) in April 2026, aiming to cut annualized operating costs by more than $500 million in the second half and reallocate resources to core areas such as AI research and development and user growth.
Q2 revenue grew 19% year-over-year to $1.599 billion, net loss narrowed from $263 million in the year-ago period to $164 million, and adjusted EBITDA jumped from $41 million to $250 million.
Fundamentals are improving, but the sustained investment required for SPECS, along with the long cycle of AR hardware transitioning from "developer tool" to "mass consumer product," is testing investors' patience.
Grok Bot User Data and the AI Narrative Shift
On the same day, another piece of news indirectly intensified market scrutiny of Snap.
Musk reposted user growth data for SpaceXAI's Grok Bot on X: for the week ending September 14, the product reached 418,000 weekly active users, up 24% from the prior week.
Grok Bot launched in mid-August, positioned not as a traditional conversational AI but as an "autonomous digital employee" capable of handling enterprise workflows including email management, database updates, invoice processing, scheduling, and software bug submission.
SpaceXAI packages it into multi-tier enterprise AI subscription plans while also offering a standalone app.
Team pricing is $120 per seat per month, and individual pricing is $200 per month, making it price-competitive among enterprise-grade AI agent products.
Notably, Grok Bot's enterprise customer service application has shown initial results — SpaceXAI says its customer service team increased ticket processing volume by 175% without adding headcount, with some issue resolution costs as low as $0.20 to $0.30.
The company also simultaneously released the Grok 4.7 model for coding and knowledge work.
The core reason this news is connected to Snap is that market attention has shifted toward AI FOMO — the fear that enterprises will fall behind others by not keeping up with AI technology development.
When Musk describes a user growth rate for an AI agent product as "we've never seen anything grow this fast," investors' tolerance for hardware narratives — especially AR glasses requiring long-cycle investment with uncertain short-term returns — correspondingly decreases.
The AI agent market is currently on the eve of an explosion. According to research firm MarketsandMarkets, the global AI agent market is projected to reach $19.3 billion in 2026 and $205.88 billion by 2033, representing a compound annual growth rate of 40.2% during the period.
Meta's Muse AI, OpenAI's Agent products, Anthropic's Claude Cowork, and SpaceXAI's Grok Bot are pushing this赛道 toward white-hot competition.
Snap's SPECS Intelligence direction aligns with this trend, but it is clearly lagging in progress.
Only a limited preview is available on iOS, and the full Mac service remains invitation-only, leaving a considerable gap before it can truly compete in AI agents.
Comments